| With the rapid development of science and technology, digitalization of image and voice with data storage and compression techniques have made great progress in recent years. Streaming media technology has touched every aspect of our lives. Network video surveillance system which is based on audio and video compressing encoding technique and network streaming media technology has presented increasingly widespread applications and showing a booming trend in our daily life, finance, insurance, intelligent transportation and other areas.In this paper, audio and video network surveillance system software based on multimedia TMS320DM368processor is designed and implemented. This system software is constituted by video/audio capture, encoding and transmission components for realizing the functions of capture, encoding, transmission and playing of the audio data and high-definition video image. In addition, this paper has done some research in audio and video synchronization and shadow removal of dynamic objects.Firstly, the research significance and practical application value has been introduced with emphasis in this paper. The overall realization scheme of the system and the system development platform building process have been given in details.Secondly, the main composition modules and their functions of the system front-end software have been elaborated. The successfully transplanted system has basic drives of video, audio, network cards, etc. Through the software design of the upper layer applications, basic functions such as audio and video data acquisition, processing and transmission have been realized. In this system, the video module uses the V4L2interface function and the audio module uses the ALSA interface function. Video data compression algorithm of H264and audio data compression algorithm of acc are used. The Data transmission module uses the open source ortp library functions.Thirdly, a detailed study has been done in this paper on the audio and video synchronization technology in video surveillance and the shadow removal technology in moving target detection. A highly efficient method is proposed for realizing audio and video synchronization based on the feedback early warning algorithm. Combining use of multi-frame average method and HSV color space algorithm is processed to remove the shadow and has achieved the expected results.Finally, the overall system functionalities have been tested.The results indicate that this system can effectively realize the functions of audio and HD video data capturing, coding and real-time display; playback of the audio and video data saved as local files can be achieved by using general players. |
